|
Descripción
|
|
|---|---|
| Nowadays, the growth of the Internet of things makes necessary to improve systems in terms of reliability, autonomy, and adaptation. Some research lines are focused on these issues to be part of new necessities. The main idea of this chapter is to go further than a wide extended communication among devices or remote control focusing on decision making of cooperative systems. Voting algorithms are widespread in this kind of applications since they allow to combine multiple outputs to generate the final solution. The simplest voting methodology is majority but, in this case, past classifications do not affect the following ones. However, weighted majority can make the system adaptive since weights are calculated according to the previous behavior of each device. Two different weighted methods are analyzed in this chapter. The first one establishes the definition of how weights have to evolve depending on the matches between the solution of each device and the final cooperative solution. In contrast, the second weighted approach estimates weights using a stochastic-based method which gives weight assignments after analyzing multiple combinations. Rewards and penalties will be different every time. Final weights are not given by a specific combination; they are calculated according to all the valid combination distribution (i.e., geometric center of all of them). The way to define the valid combinations will determine the system reliability. Additionally, in order to verify the performance of each method, a case of use is also presented. Results demonstrate the adaptation of both methods and how the system reliability is also improved comparing to the simple majority solution. | |
|
Internacional
|
Si |
|
DOI
|
|
|
Edición del Libro
|
|
|
Editorial del Libro
|
Springer |
|
ISBN
|
978-981-13-1768-2 |
|
Serie
|
|
|
Título del Libro
|
Smart Grids and Their Communication System |
|
Desde página
|
499 |
|
Hasta página
|
529 |